| Term | Definition |
|---|
| Traffic Source/Medium | Where users came from (e.g., google / organic,
linkedin.com / referral) |
| Engaged Sessions | Sessions that lasted longer than 10s, had a conversion event, or 2+ views |
| Signup | User completes signup form |
| Activate | User completes key onboarding step (e.g., email verify, login) |
| Setup Complete | User finishes full onboarding or setup funnel |
| Ebook CTA | User clicks the call-to-action for an ebook |
| Ebook Submit | User successfully submits/downloads the ebook |
| DAU/MAU Ratio | Stickiness metric (daily active users ÷ monthly active users) |
Provide valuable insights to Productfolio.com by identifying which media channels and content types drive the most traffic and conversions.
To meet this objective, we explored four core areas:
Traffic Acquisition
Landing Page Funnel Analysis
Site Engagement & Pageviews
Ebook Funnel Insights
Control Variables
library(readr)## Warning: package 'readr' was built under R version 4.4.2traffic <- read_csv("Traffic_acquisition_Clean.csv")## Rows: 481 Columns: 9
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): source_medium
## dbl (8): sessions, engaged_sessions, engagement_rate, avg_engagement_time, e...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.library(dplyr)## Warning: package 'dplyr' was built under R version 4.4.2##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(ggplot2)## Warning: package 'ggplot2' was built under R version 4.4.2top_sources <- traffic %>%
arrange(desc(sessions)) %>%
slice_head(n = 10)ggplot(top_sources, aes(x = reorder(source_medium, sessions), y = sessions)) +
geom_col(fill = "#69b3a2") +
coord_flip() +
labs(
title = "Top 10 Traffic Sources by Sessions",
x = "Source / Medium",
y = "Sessions"
) +
theme_minimal()
Observations:
Google / organic is the top traffic source. It significantly outperforms the other channels.
The Direct traffic section are users arriving without a known source. Direct traffic ranks second, but with noticeably fewer sessions than Google.
Other sources like bing, referrals, and Google/ CPC contribute minimal traffic in comparison.
Google/CPC: Cost per click advertisements.
Interpretations:
Productfolio is gaining significant traction primarily through Google Organic Search. Their SEO strategy works. There should be a continued investment in content and keyword optimization.
The high percentage of direct traffic may indicate returning users or unattributed sessions (e.g., from untagged links); UTM tracking can help clarify and capitalize on these visits.
Referral traffic is limited in sessions. There might be potential to expand presence through partnerships or backlink outreach.
top_engaged <- traffic %>%
arrange(desc(engaged_sessions)) %>%
slice_head(n = 10)ggplot(top_engaged, aes(x = reorder(source_medium, engaged_sessions), y = engaged_sessions)) +
geom_col(fill = "#FFA07A") +
coord_flip() +
labs(
title = "Top 10 Sources by Engaged Sessions",
x = "Source / Medium",
y = "Engaged Sessions"
) +
theme_minimal()
Observations:
Google / organic has the highest, most high quality engagement.
Direct traffic also generates a moderate number of engaged sessions but at a much lower scale than Google.
trafficpeak.io brought 639 sessions and 100% of them were engaged sessions
Interpretations:
Google organic highlights that users who arrive via search are numerous and highly engaged.
Direct traffic shows decent interaction but doesn’t have as much success with conversions. This suggests that both loyal users and untagged referrals are taken place here.
The low engagement from paid channels (e.g., Google / CPC) and niche referrals may indicate a need for a reallocation of resources.
Reach out to trafficpeak.io and collaborate, promote more content there, learn what audience they attract
top_events <- traffic %>%
filter(!grepl("\\(not set\\)", source_medium)) %>%
arrange(desc(events_per_session)) %>%
slice_head(n = 10)top_events <- traffic %>%
filter(!grepl("\\(not set\\)", source_medium)) %>%
arrange(desc(events_per_session)) %>%
slice_head(n = 10)
ggplot(top_events, aes(x = reorder(source_medium, events_per_session), y = events_per_session)) +
geom_col(fill = "#9370DB") +
coord_flip() +
labs(
title = "Top 10 Sources by Events Per Session (Filtered)",
x = "Source / Medium",
y = "Avg. Events per Session"
) +
theme_minimal()
inobram.atlassian.net / referral
leads significantly, averaging over 22 events per
session.
Most sources are referral-based.
Google PPC ranking 5th in this list ndicates that while paid search brings traffic, it may not match the organic engagement depth of referrals.
Referral traffic — particularly from niche or tech-aligned sources — ususally yields more high engagement users. This is probably because the users are already aligned with Productfolio’s offerings before arriving.
High event counts per session imply that these users are actively exploring, clicking, or triggering multiple events, which could indicate interest in product features, content, or conversions.
We can prioritize partnerships and investigate what techniques they’re using, especially with platforms like Poptin and Product Hunt.
Which landing pages are the most efficient at converting visitors into signups?
Which landing pages are most effective at converting users through the funnel (signup → activation → setup completion)
Observations:
/) drives the highest number of signups,
totaling over 60. Meanwhile,
/21-product-management-frameworks and
/what-is-a-digital-product follow, each generating 20–30
signups.Interpretations:
Educational and informative content such as product frameworks and digital product explanations appear highly effective in motivating users to sign up.
Targeted pages like /signup may attract users who
were already intending to convert. Thiss suggests strong value in having
direct, accessible signup points alongside content-driven
pages.
Observations:
/definition-of-done leads all landing pages with a
signup rate above 0.35. Meanwhile,
/cal-poly/cis/roadmap follows with a signup rate around
0.30.
The last few pages all show moderate signup rates between 0.07–0.09.
Interpretation:
/definition-of-done and
/cal-poly/cis/roadmap are very effective at having users
initiate signups. This may be due to strong messaging, relevant content,
or high-intent traffic.
Observations:
/definition-of-done and
/cal-poly/cis/roadmap still rank top 2 in activation,
indicating that the user continued engagement beyond the
signup.
Activation rates remain consistent with signup trends. This shows that the customers who signed up are also completing key next steps like email verification or onboarding.
Lower-ranked pages show a drop-off in activation despite decent signup rates.
Interpretation:
The top pages not only attract but retain user momentum through the funnel.
Activation strength makes sure that the signup that was initially done wasnt accidental or casual. Activation rates show users who are genuinely engaged and willing to move forward with the process.
Observations:
/definition-of-done dominates again with
/Beta-testingreplacing
/cal-poly/cis/roadmap.
Pages like /ebook-downloads and
/product-portal appear in the top 5 here, even though they
weren’t strong performers in signups or activation.
Interpretation:
Setup completion indicates full onboarding, and
/definition-of-done is most efficient across the entire
funnel.
Pages like /ebook-downloads improved
visibility.
How does average user engagement evolve across the year?
Are the most visited pages also the most engaging?
How frequently do users return to engage with the site (DAU/MAU)?
Observations:
The average engagement time per active was relatively stable across the year. It usually ranged between 0.5 to 1 minute on most days.
Notable irregular and isolated spikes would occur sometimes, with the longest times reaching 4 to 8 minutes.
Interpretation:
There appears to be a stable user base behavior with occasional peaks likely driven by specific events. (We don’t know for sure, further investigation should be conducted.)
Understanding what caused these spikes (e.g., blog releases, email campaigns, feature updates) could help Productfolio optimize content strategy to replicate these high-engagement moments more frequently throughout the year.
Observations:
The majority of these pages are stacked in the bottom-left quadrant, indicating low views and low engagement time.
There are some outlier pages that have very high pageviews (e.g., over 20,000), but their engagement times are only modestly higher than the average.
Some pages with moderate engagement time (1–1.5 minutes) received relatively low views. This indicates that the pages are highly engaging but are underexposed.
Interpretation:
There is no clear correlation between total page views and average engagement time — pages that attract high traffic don’t necessarily keep users engaged longer.
Likewise, traffic volume does not guarantee quality engagement, and highly engaging content may be getting overlooked due to limited exposure.
This is an opportunity to amplify lesser-known but engaging content — potentially through better internal linking, SEO, or promotional strategies.
Additionally, evaluating why top-viewed pages aren’t driving higher engagement.
Observation
The top three pages by engagement time (orange)
— /janet/autozon/dashboards/a/sa/dashboard,
/ssss/wwww/prioritize, and
/csa/sa/dashboard
In contrast, the most visited pages (green) —
/21-product-management-frameworks/,
/what-is-a-digital-product/, and
/ice-scoring/
Interpretation
Not all popular pages are engaging, and not all engaging pages are popular.
Observations:
The DAU/MAU ratio fluctuates between 0.03 and 0.05 for most of the year.
There are occasional spikes (e.g., near days ~50, ~140, ~250, and ~365) where the stickiness ratio briefly increases, peaking around 0.11 and again near the industry benchmark of 0.15.
Despite the short-term peaks, the majority of values remain well below the industry average of 0.15.
Interpretation:
Productfolio’s stickiness is low. This means that while users may visit the site, they are not returning frequently within a month.
The low DAU/MAU ratio (~0.04 average) indicates a need to strengthen re-engagement strategies to keep users coming back.
Which channels lead to the most ebook submissions?
How does ebook engagement compare to other key events on the site?
Observations:
Google is the top-performing channel by a large margin, contributing 89 ebook submissions.
Direct traffic follows with 26 submissions, indicating users may be accessing the site by typing in the URL or via untagged links.
Interpretations:
Search engines (especially Google) play a dominant role in driving ebook engagement, highlighting the importance of strong SEO and discoverability of ebook-related content.
The significant volume from direct traffic may suggest returning users or links in emails/DMs without tracking parameters – we should consider improving UTM tagging for better attribution.
Platforms like ChatGPT.com are beginning to show traction and like LinkedIn and Medium underperform despite being content-focused platforms. This may signal an opportunity to boost ebook visibility on these networks via targeted campaigns or reposting.
Observations:
The “Ebook CTA” ranks 5th
overall in total user interaction, just behind foundational events like
first_visit, session_start, and
user_engagement.
Events directly related to the ebook journey —
form_start,
form_submit,
ebook_submit, and
file_download — follow closely in the
ranking.
The top 4 events are broad, universal events that most users trigger during any visit. Good news for ebook downloads.
Interpretations:
Ebook engagement is one of the most triggered intentional user actions on the site, placing it in the top half of key site event. This is a strong signal of its relevance and value.
The full ebook funnel appears healthy, with high conversion from click to form start to submission,.
Ebooks drive meaningful engagement and action.
Google (organic search) is the leading source of both sessions and engaged sessions, confirming the strength of SEO as a top acquisition channel.
Referral sources like
inobram.atlassian.net and producthunt.com
drive the highest per-session interaction, highlighting the value of
strategic partnerships and placements.
High-performing landing pages like
/definition-of-done and /cal-poly/cis/roadmap
not only attract visitors but successfully guide them through signups,
activation, and full setup completion.
While overall stickiness (DAU/MAU) remains below the industry benchmark of 0.15, patterns suggest spikes in return visits during key content releases or campaign pushes.
Ebooks stand out as one of the top conversion mechanisms. Ebook-related events are among the highest-performing site actions, especially when traffic is driven by Google.
Double down on SEO and high-converting content like definitions, frameworks, and comparison pages — they drive both traffic and funnel progression.
Prioritize referral partnerships and content syndication opportunities with domains that show high engagement per session.
Continue promoting ebooks, and layer on email or product onboarding flows for users who engage with them.
Improve attribution on (direct) and
(not set) traffic with better UTM tagging and campaign
tracking.
Explore stickiness improvements through in-app reminders, content updates, or email nurturing to encourage return visits.